Recent publications of the Institute of Medicine have focused national attention on issues of patient safety and have suggested that information technology may hold the potential to improve patient safety, by reducing errors and by creating organizational efficiencies. The Agency for Healthcare Research and Quality has recommended, as a strategy to improve patient safety, the implementation of computerized prescriber order entry (CPOE) systems with built-in clinical decision support (CDS). To date, this recommended strategy has been implemented and evaluated primarily in inpatient and academic: settings; less frequently in the ambulatory setting of a community-based, integrated health-system. The primary objectives of this project are to implement a CPOE system with CDS capabilities in an ambulatory, community-based, integrated health-system and to evaluate the impact of the system both internally, on organizational processes and human factors; and externally, on patient safety as measured by medication errors and adverse drug events. This three-year project has two specific aims. Aim 1 is to implement the ambulatory CPOE/ CDS system in all thirteen sites of The Everett Clinic, a community-based integrated health-system in the north Puget Sound region. We will document lessons learned, and evaluate the impact of the system on organizational workload processes and on the behavior of the health professionals who use the system. In Aim 2 we will evaluate the impact of the ambulatory CPOE/ CDS system on patient safety, by characterizing the epidemiology of medication errors pre- and post-implementation; and linking these errors to adverse events. While the implementation will take place in all thirteen sites, portions of the evaluation will take place in both a family practice and a pediatric clinic, with two matched clinics serving as control groups. In Aim 1, strategies used for, and lessons learned in achieving success will be documented. Workload metrics will be captured using both retrospective reporting mechanisms and a time and motion analysis. The relationship between information technology and human factors will be captured via informant interviews, focus groups and administration of a survey instrument to health professionals. In Aim 2, 10,000 prescriptions will be retrospectively evaluated to identify and characterize medication errors and potential errors, using definitions and severity levels defined a priori. Errors will be linked to adverse events that cause emergency department or hospital admissions. All evaluations will compare data pre-to post-implementation. Resultant manuscripts will focus on generalizability so that external stakeholders can benefit, from our work. This will be one of the first projects wherein implementation of a CPOE/CDS system in an ambulatory community-based setting will be evaluated and results disseminated in a structured and systematic way. [unreadable] [unreadable]